The goal of this research application is to define cellular interactions that guide T lymphocyte development. Experiments performed in the previous funding period have focused on recognition events involved in positive selection, and have established that, 1) positive selection occurs on conventional peptide/MHC complexes, that 2) peptides presented in the thymus are involved in shaping the T cell repertoire, and yet that 3) recognition of positive selection is promiscuous. In the current application, it is argued that recognition events of positive selection can only be understood if the natural ligands of positive selection are known. Yet, for conventional T cell receptor transgenic mice, it will be impossible to identify a single or few naturally-occurring positively selecting peptide(s), partly because positive selection is promiscuous in nature, and it is possible that positive selection occurs on a "gemisch" of different peptides. Therefore, an experimental model will be specifically designed to limit the number of candidate positively-selecting peptides. The studies in this application are directed toward generating a novel T cell receptor transgenic mouse in which the number of positively selecting peptides is limited to thirteen peptides, of defined sequence. Identification of the natural ligand(s) of positive selection in this system will make it possible to test the validity of the current hypothesis of positive selection. The final goal of the application is the definitive characterization of recognition events of positive selection.